Interlaboratory comparison of categorical characteristics of a substance, material, or object

Ilya Kuselman, Tamar Gadrich, Francesca R. Pennecchi, D. Brynn Hibbert, Anastasia A. Semenova, Angelique Botha

Abstract

When a reference value for the measurand in an interlaboratory study is unknown, the laboratory results may be used to estimate/build a consensus value instead of a reference value. Since no algebraic operations and mathematical functions exist among categorical (nominal and ordinal) values, a numerical consensus value cannot be formulated. Consensus of responses of experts of different laboratories participating in an interlaboratory comparison, classifying a substance, material, or object according to its nominal and ordinal characteristics, could be interpreted as the degree to which the experts agree. Two-way factorial analysis of variation of nominal variables CATANOVA and of ordinal variables ORDANOVA answer the question ‘is a consensus of participating laboratories achieved or not?’ The answer is based on testing hypotheses about homogeneity of the between- and within-laboratory variation components, as well as the variation components caused by other factors under study.

Download

IMEKO-TC8-11-24-2025-029.pdf

DOI

10.21014/tc8-2025.029

IMEKO TC

TC8 - Traceability in Metrology

Back to the proceedings